[FFmpeg-devel] [PATCH] MMX implementation of VC-1 inverse transforms

Balatoni Denes dbalatoni
Mon Jan 14 23:10:43 CET 2008


Monday 14 January 2008 21:07-kor M?ns Rullg?rd ezt ?rta:
> Balatoni Denes <dbalatoni at interware.hu> writes:
> > Figuring out how the other IDCTs work is not so simple task as you make
> > it look like. For example the simple_idct is the fastest and most
> > accurate, but it is quite involved (some crazy reordering imho), I for
> > one couldn't figure out how it works during the time I spent on it.
> Realising how it implements an IDCT is indeed non-trivial.  Writing a
> SIMD version is still a fairly straight-forward task, though.

Obviously I was not talking about the algorithm (which is given, and well 
documented), but the SIMD version made from it in the case of simple_idct. 
Yes, I failed to understand the simd version. I am that big a lamer.


More information about the ffmpeg-devel mailing list